Path: utzoo!attcan!uunet!husc6!bloom-beacon!mit-eddie!uw-beaver!cornell!rochester!udel!princeton!phoenix!pucc!EGNILGES
From: EGNILGES@pucc.Princeton.EDU (Ed Nilges)
Newsgroups: comp.software-eng
Subject: Re: Implementing algorithms
Message-ID: <5271@pucc.Princeton.EDU>
Date: 16 May 88 16:04:11 GMT
References: <3278@enea.se>
Reply-To: EGNILGES@pucc.Princeton.EDU
Organization: Princeton University, NJ
Lines: 14
Disclaimer: Author bears full responsibility for contents of this article

In article <3278@enea.se>, sommar@enea.se (Erland Sommarskog) writes:
 
>Knuth does not discuss this problem here,
>so we have to think for ourselves.
 
Nothing wrong with that!  The reason for using algorithms in books (or
canned software) is not to prevent thinking, but to build a better
product, and give the maintenance programmer a resource.
 
I typically have to manually translate Knuth's algorithms into a
structured style. This is not a waste of time, since I understand
the algorithm if I have structured it.